Increasing the productivity of the rice-wheat (RW) system in north-west India is critical for the food security of India. However, yields are stagnating or declining, and the rate of groundwater use is not sustainable. Many improved technologies are under development for RW systems, with multiple objectives including increased production, improved soil fertility, greater input use efficiency, reduced environmental pollution, and higher profitability for farmers. There are large reductions in irrigation amount with many of these technologies compared with conventional practice, such as laser land leveling, alternate wetting and drying (AWD) water management in rice, delayed rice transplanting, shorter duration rice varieties, zero till wheat, raised beds, and replacing rice with other crops. However, the nature of the irrigation water savings has seldom been determined. It is often likely to be due to reduced deep drainage, with little effect on evapotranspiration (ET). Reducing deep drainage has major benefits, including reduced energy consumption to pump groundwater, nutrient loss by leaching, and groundwater pollution. The impacts of alternative technologies on deep drainage (and thus on irrigation water savings) vary greatly depending on site conditions, especially soil permeability, depth to the watertable, and water management. More than 90% of the major RW areas in north-west India are irrigated using groundwater. Here, reducing deep drainage will not "save water" nor reduce the rate of decline of the watertable. In these regions, it is critical that technologies that decrease ET and increase the amount of crop produced per amount of water lost as ET (i.e., crop water productivity, WP ET) are implemented. The best technologies for achieving this are delaying rice transplanting and short duration rice varieties. The potential for replacing rice with other crops with lower ET is less clear.
